An application of mixture distributions in modelization of length of hospital stay

Length of hospital stay (LOS) is an important indicator of the hospital activity and management of health care. The skewness exhibited by this variable poses problems in statistical modeling. The aim of this work is to model the variable LOS within diagnosis-related groups (DRG) through finite mixtures of distributions. A mixture of the union of Gamma, Weibull and Lognormal families is used in the model, instead of a mixture of a unique distribution family. Some theoretical questions regarding the model, such as the identifiability and study of asymptotic properties of ML estimators, are analyzed. The EM algorithm is proposed for performing these estimators. Finally, this new proposed model is illustrated by using data from different DRGs.

[1]  Charles E. Antle,et al.  Estimation of Parameters in the Weibdl Distribution , 1967 .

[2]  David Durand,et al.  Aids for Fitting the Gamma Distribution by Maximum Likelihood , 1960 .

[3]  R. Redner,et al.  Mixture densities, maximum likelihood, and the EM algorithm , 1984 .

[4]  A. Lee,et al.  Public versus private hospital maternity length of stay: a gamma mixture modelling approach , 2002, Health services management research.

[5]  J. M. Muñoz-Pichardo,et al.  On the consistency of MLE in finite mixture models of exponential families , 2007 .

[6]  Charles E. Antle,et al.  Estimation of Parameters in the Weibull Distribution , 1967 .

[7]  J. Muñoz-Pichardo,et al.  Consistency of Maximum Likelihood Estimators in Finite Mixture Models of the Union of 𝒲-type Families , 2005 .

[8]  A Marazzi,et al.  Fitting the distributions of length of stay by parametric models. , 1998, Medical care.

[9]  J. M. Muñoz-Pichardo,et al.  A new condition for identifiability of finite mixture distributions , 2006 .

[10]  Andy H. Lee,et al.  Mixture distribution analysis of length of hospital stay for efficient funding , 1999 .

[11]  A. F. Smith,et al.  Statistical analysis of finite mixture distributions , 1986 .

[12]  J. Henna.  Examples of Identifiable Mixture , 1994 .

[13]  R. Fetter,et al.  Case mix definition by diagnosis-related groups. , 1980, Medical care.

[14]  L Dusserre,et al.  High cost factors for leukaemia and lymphoma patients: a new analysis of costs within these diagnosis related groups. , 1999, Journal of epidemiology and community health.

[15]  H. Teicher.  Identifiability of Mixtures of Product Measures , 1967 .

[16]  R. Jennrich,et al.  Conjugate Gradient Acceleration of the EM Algorithm , 1993 .

[17]  M. Silberbach,et al.  Predicting hospital charge and length of stay for congenital heart disease surgery. , 1993, The American journal of cardiology.

[18]  K. Chanda.  A NOTE ON THE CONSISTENCY AND MAXIMA OF THE ROOTS OF LIKELIHOOD EQUATIONS , 1954 .

[19]  K. E. Ahmad.  Identifiability of finite mixtures using a new transform , 1988 .

[20]  I McColl,et al.  Firm, patient, and process variables associated with length of stay in four diseases , 1978, British medical journal.

[21]  O. Barndorff-Nielsen.  Identifiability of mixtures of exponential families , 1965 .

[22]  D. Quade,et al.  A method for constructing case-mix indexes, with application to hospital length of stay. , 1986, Health services research.

[23]  G. McLachlan,et al.  The EM algorithm and extensions , 1996 .

[24]  M. Cleman,et al.  Length of hospital stay and complications after percutaneous transluminal coronary angioplasty. Clinical and procedural predictors. Heparin Registry Investigators. , 1995, Circulation.

[25]  Andy H. Lee,et al.  Determinants of Maternity Length of Stay: A Gamma Mixture Risk-Adjusted Model , 2001, Health care management science.

[26]  L Dusserre,et al.  Modeling of high-cost patient distribution within renal failure diagnosis related group. , 1999, Journal of clinical epidemiology.

[27]  Dimitris Karlis,et al.  Choosing Initial Values for the EM Algorithm for Finite Mixtures , 2003, Comput. Stat. Data Anal..